Geometry-based multiplication correction for passive neutron coincidence assay of materials with variable and unknown (alpha,n) neutron rates

摘要

We have studied the problem of assaying impure plutonium-bearing materials using passive neutron coincidence counting. We have developed a technique to analyze neutron coincidence data from impure plutonium samples that uses the bulk geometry of the sample to correct for multiplication in samples for which the ((alpha),n) neutron production rate is unknown. This technique can be applied to any impure plutonium-bearing material whose matrix constituents are approximately constant, whose self-multiplication is low to moderate, whose plutonium isotopic composition is known and not substantially varying, and whose bulk geometry is measurable or can be derived. This technique requires a set of reference materials that have well-characterized plutonium contents. These reference materials are measured once to derive a calibration that is specific to the neutron detector and the material. The technique has been applied to molten salt extraction residues, PuF(sub 4) samples that have a variable salt matrix, and impure plutonium oxide samples. It is also applied to pure plutonium oxide samples for comparison. Assays accurate to 4% (1 (sigma)) were obtained for impure samples measured in a High-Level Neutron Coincidence Counter II. The effects on the technique of variations in neutron detector efficiency with energy and the effects of neutron capture in the sample are discussed.